pts. All pts were assessed post-op every 3 months for the 1st year, every 6 months for the 2nd year and yearly thereafter. Post-op UC recovery was defined as the absence of any protection device (no pads). Pts were divided into 3 pre-op risk groups: low (PSA10 ng/ml, cT1, BxGS 6), high (cT3 or BxGS 8–10 or PSA 20 ng/ml) and intermediate risk (all the remaining patients). Kaplan-Meier curves assessed the time to UC recovery in the overall patient population as well as according to AdjRT administration. The association between AdjRT and UC recovery was assessed in univariable and multivariable Cox regression analyses after accounting for age, pre-op risk groups, NS status and year of surgery. RESULTS: Pre-op risk groups allowed to identify 528 (47.0%), 460 (41.0%) and 135 (12.0%) pts in the low, intermediate and high risk group, respectively. The 1 and 2-years UC recovery rates were signif- icantly higher for pts not treated with AdjRT, as compared to pts receiving AdjRT (p0.001). When pts were stratified according to D’Amico risk groups, AdjRT did not impact on the rate of UC recovery in low risk pts(Log rank p=0.1), while it significantly affected the rate of UC recovery in intermediate (1yr 67% vs 43%; 2yr 74% vs 47%; p0.001) and high risk pts(1yr 47% vs 32%; 2yr 52% vs 39%; p=0.042). These data were confirmed at multivariable Cox analyses where AdjRT was independently associated with a lower rate of UC recovery (p0.001), even after accounting for patient age, clinical oncologic characteristics, NS status and year of surgery. Pts not receiving AdjRT had a 1.6 fold higher probability of recovering full continence after surgery. CONCLUSIONS: Patients treated with AdjRT have a de- creased probability of achieving full continence (no pads) after RP. This results should be taken in account when RP and AdjRT are considered as a treatment option in pts with intermediate or high risk PCa. Source of Funding: None 724 TESTOSTERONE RECOVERY IN PATIENTS WITH HIGH RISK PROSTATE CANCER TREATED WITH 36 VS 18 MONTHS OF ANDROGEN BLOCKADE AND PELVIC IRRADIATION Abdenour Nabid*, Nathalie Carrier, Sherbrooke, Canada; Andre ´ -Guy Martin, Quebec, Canada; Jean-Paul Bahary, Luis Souhami, Montreal, Canada; Francois Vincent, Trois-Rivie ` res, Canada; Sylvie Vass, Chicoutimi, Canada; Boris Bahoric, Montreal, Canada; Robert Archambault, Gatineau, Canada; Ce ´ line Lemaire, Montreal, Canada; Sabrina Selmani, Rimouski, Canada INTRODUCTION AND OBJECTIVES: The purpose of the pres- ent analysis is to evaluate testosterone recovery in patients with high risk prostate cancer who received androgen blockade (AB) and radiother- apy (RT) in a prospective randomized trial (PCS IV ClinicalTrials.gov, #NCT00223171). METHODS: PCS IV randomized patients with high risk prostate cancer to 36 (arm 1) vs 18 (arm 2) months of AB plus pelvic RT. Serum testosterone was measured at baseline, then at each visit. 47 patients (arm 1) and 38 patients (arm 2) were excluded from the analysis (4 and 5 no testosterone measured, 25 and 18 no baseline testosterone, 18 and 15 only baseline testosterone, respectively). For patients who developed biochemical failure, testosterone data were collected until the beginning of a second course of AB. RESULTS: From October 2000 to January 2008, 310 patients were randomized to arm 1 and 320 patients to arm 2. There are no significant differences in pre treatment characteristics between the 2 arms except for the initial testosterone level (normal level in 77.9% of patients in arm 1 and 70.6% in arm 2, p=0.049). The median age, initial PSA and Gleason score for arm 1 and 2 were 70.0 vs 69.6 years, 16.0 vs 15.2 ng/ml and 8 vs 8, respectively. The clinical stage distribution was similar between arms (25.1% T1c, 21.3% T2a, 29.4% T2b and 24.0% T3). All patients developed a castrate status post AB. With a median follow-up of 66 months for arm 1 and 55.3 months for arm 2 (p=0.02), the proportion of patients who recovered normal testosterone is higher in arm 2 (41.5% vs 25%, p0.001) and in patients with initial normal vs abnormal testosterone (arm 1: 27.8% (57/205) vs 15.5% (9/58), p=0.057; arm 2: 48.7% (97/199) vs 24.1% (20/83), p0.001). Time to testosterone recovery (25th percentile) was 60 months (55.5– 64.5) for arm 1 vs 36 months (33.8 –38.2) for arm 2, (p  0.001). Initial testosterone level did not correlate with patients’ age (arm 1: r=0.088, arm 2: r=-0.018, all p0.05). A Cox regression analysis showed that younger age, initial normal testosterone level and AB duration are significant factors for testosterone recovery (Hazard Ratio: age 0.95, initial normal testosterone level 2.16 and 18 months of AB 2.97, all p0.001). CONCLUSIONS: In patients with high risk prostate cancer treated with 36 vs 18 months of AB and RT, 25% and 42% will respectively recover normal testosterone with a shorter time in arm 2. Younger patients and those with initial normal testosterone have a higher recovery rate. These data can be useful when discussing side effects of long term AB with patients and may have a significant impact in outcomes. Source of Funding: AstraZeneca Pharmaceuticals grant Prostate Cancer: Basic Research Moderated Poster 24 Monday, May 16, 2011 8:00 AM-10:00 AM 725 PROSTATIC DISEASES IN THE ELDERLY MEN: GLANDULAR MICROENVIRONMENT SIGNALING OF IGFR-1, STEROID HORMONE RECEPTORS, MATRIX METALLOPROTEINASES AND DYSTROGLYCANS Vale ´ ria Quitete*, Amanda Hetzl, Wagner Fa ´ varo, Ubirajara Ferreira, Athanase Billis, Campinas, Brazil INTRODUCTION AND OBJECTIVES: Senescence is a deter- mining factor for the occurrence of morphological changes in the prostate. Thus, the objective of this work was to characterize and correlate the  and -dystroglycans (DG, DG), androgen receptor (AR),  and -estrogen receptors (ER, ER), matrix metalloproteinases 2 and 9 (MMP-2, MMP-9) and insulin-like growth factor receptor (IGFR-1) reactivities in both stroma and epithelium compartments of elderly men with Benign Prostatic Hyperplasia (BPH), Prostatic Intraepithelial Neoplasia (HGPIN) and Prostatic Cancer (PC). METHODS: Sixty prostatic peripheral zone samples from 60 –90 year old patients were divided into 4 groups: The Standard Group (no lesions); The HGPIN Group; The PC and The BPH Group and analyzed by means of immunohistochemistry and Western Blotting analyses. RESULTS: The results showed increased IGFR-1, MMP-2 and MMP-9 protein levels in the PC and PIN groups in relation to other groups. Decreased DG and DG protein levels were verified in the PC and PIN groups in relation to the BPH and Standard groups. Intensified AR immunoreactivity was verified in the epithelial compartment in all studied groups. ER immunoreactivity was more intense in the epithe- lial compartment in the PC and PIN groups than in the other groups. ER immunoreactivity was weaker in the epithelial compartment of the PIN and PC groups than in the BPH and Standard groups. CONCLUSIONS: Abnormal DG, DG, IGFR-1, MMP-2 and MMP-9 levels certainly compromised the glandular epithelial-stromal interaction in the senescence. The differential steroid hormone receptor reactivities in the glandular lesions in both prostatic compartments indicated different paracrine signals to the dynamics of the prostate and pointed out the importance of estrogenic pathways in the activation of these changes. Also, these findings indicated that there was a direct Vol. 185, No. 4S, Supplement, Monday, May 16, 2011 THE JOURNAL OF UROLOGY e291